Is Transplantation for CML Dead?

Chronic myeloid leukemia (CML) is hardly a public health menace, yet it embodies the "bench to bedside" mantra that drives translational medicine. The discovery of the Philadelphia chromosome, the juxtaposition of the BCR and ABL genes to form the t(9;22), fostered understanding of aberrant signal transduction, targets for minimal residual disease monitoring by PCR, and ultimately "targeted" therapy by tyrosine kinase inhibitors (for example, imatinib). Prior to the advent of imatinib, biologic therapy with interferon was shown to lengthen the natural history of CML and was deemed to be appropriate therapy for patients without an allogeneic transplant option. Transplantation has been considered a "curative" therapy for CML, but not without the trade-off of considerable morbidity and mortality.

How to treat newly diagnosed chronic phase CML has changed dramatically with imatinib.

• Imatinib is remarkably effective for patients treated in chronic phase, as greater than 75 percent of patients obtain a complete cytogenetic remission (CCyR). In the IRIS trial, approximately 70 percent of cases remain in a CCyR at five years of follow-up¹ . Treatment of advanced phase (accelerated of blast disease) is associated with much poorer outcomes.

• Allogeneic transplantation is generally associated with 10-year survivals of 70 percent or better for patients in chronic phase, but survival likewise falls in accelerated or blast-phase disease^2-4 .

• Relapse occurs for chronic-phase patients treated with imatinib, but outcome can be effectively monitored by sensitive RT-PCR assays^5-7 .

Current recommendations from advisory panels, such as the European Leukemia Net and the National Cancer Care Alliance, state that all chronic phase patients start on imatinib therapy, but allow for consideration of transplantation based on the patient’s age, preference, and response to initial imatinib^8,9 . The tacit assumption is that, given the excellent results of both imatinib and transplantation, a contemporary randomized trial comparing the methods would be very unlikely.

Given this background, the paper by Helmann et al. bears consideration¹⁰ . This trial looked at 621 patients with chronic phase CML; of these, 354 were considered eligible for transplantation and "biologically randomized" based on the availability of a related donor. Of the 123 patients who received a transplant, the 10-year estimate of survival was 53 percent. Those 219 patients without a related donor were treated with interferon until imatinib became available later in the trial. Imatinib was then offered to patients whose disease did not respond to interferon. The 10-year estimate of survival in this group was 52 percent. The survival curves of these two groups are shown in the figure above and show the not-surprising results: the transplant group suffers a higher early mortality, with a relative flattening of the survival curves, whereas group 2 has a better early outcome, with the survival curves continuing to drop. The cross-over of the curves comes at eight years. A statistical analysis showed that the survival rate of group 2 was significantly superior (though marginally, with a p=0.049). The authors conclude that transplantation can no longer be recommended as first-line therapy.

The undertaking of such a study is Herculean, but the conclusions deserve some scrutiny. First, the survival experience for group 2 was a blend of those who received interferon and (in some cases) imatinib, and those cases that went on to an unrelated transplant when a donor became available. The survival of those 122 patients in group 2 who only received interferon/imatinib was 50 percent; of the 97 patients who went on to receive an unrelated transplant, 69 percent survived. Moreover, it may be hard to translate this study to contemporary practice. Given the remarkable success of imatinib as first-line therapy, it would be expected that the survival curves for patients treated with imatinib up front would be superior to that of the interferon/imatinib group in this study. In addition, most transplant centers would be disappointed with the survival statistics in the transplant arm of this study of long-term outcomes (~50 percent).

Other studies have attempted to address the issue of up-front imatinib versus transplantation. One such study examined published survival data for patients who received a transplant and imatinib, modeling survival and life expectancy based on age, gender, and treatment strategies¹¹ . In most models, survival and life expectancy in the newly diagnosed chronic phase cases that initially received imatinib was superior compared with patients who initially received a transplant. The superiority of imatinib was greatest under the conditions of the low progression rate, older age, comparison to the CIBMTR transplant experience, and unrelated donor transplants. Transplantation appeared to be similar or somewhat superior to imatinib when one assumed better transplant outcomes, younger patient age, and higher imatinib relapse rates.